GHI overestimation



I was comparing hourly GHI for Trinidad with ground based measurements and I found that the PSM data is approximately 20% larger than the ground based measurements for daily averaged values. I could not find any references in the literature with a similar effect for the PSM (they give a much smaller overestimation on cloudy days), but I did find a similar overestimation for California in the SUNY model.

Has any comparison been done with a similar overestimation? If so would you be able to provide a reference? Is it possible that the PSM has an overestimation for Trinidad or is it more likely that the ground based measurements have a systematic error?

Thank you